j 
: 
i 
POSITION OF THE GENERA HYDROCOTYLE, ETC. 279 
and yet how different. is its very look from all the most typical Um- 
Uifere! None of the other European Umbellifera (for the present 
I will not mention those, of other countries) have genuine stipules 
and peltate leaves, and few such an imperfect umbel as Hydrocotyle 
has. The characters of the most typical Uméellifere the genus does 
not possess. Its fruit is didynamous, it is true, but the two carpels 
do not separate from the carpopod, nor are they vittate. To this must 
be added another highly important. character. The eestivation of the 
corolla, though described by all botanists as imbricate, is nevertheless 
truly valvate.. The unanimity with which this later point, was insisted 
upon by all the works consulted, made me anxious to have my observa- 
tions confirmed by others, and I am glad to be able to add that Messrs. 
Bennett, Carruthers, and Newbould, who saw a bud under very high 
microscopic power, were unanimous in declaring the sestivation truly 
valvate. Hydrocotyle is in fact no Umbellifera at all, but belongs to 
the same Order as Hedera Helix, especially that group which has pel- 
tate or palmate leaves and stipules.* 
The distinctive characters assigned by authors to Umbellifere and 
Araliacee break down when applied to the whole of the two Orders as 
they now stand, and it will be necessary to search for new ones which 
shall interfere least with the true limits of these two most natural of 
Natural Orders. This can best be effected, I think, by relying upon 
the wstivation of the corolla for that purpose. Restrict the name 
Umbellifere to all plants having a truly imbricate or an involute esti- 
vation, and that of Araliacee or Hederacee to all having a valvate 
or quincuncial one. -Horsfieldia, a shrubby, spiny plant, having a 
truly valvate corolla, and until now retained in Umbellifere, has been 
regarded as weakening the character derivable from the æstivation ; but 
with all. due deference to the opinion of two eminent. botanists who 
placed it there, I cannot regard Horsfieldia as a true Umbellifera ; m- 
‘deed, I have not yet been able to find any generie differences between 
it and Echinepanaz. The latter, having the same habit, probably is a 
congener of Horsfieldia ; and about its Natural Order there has never 
‘been a shadow of doubt, it being referred by Smith to Panaz (under 
* The i i hitish, and very much resemble coffee- 
pollen grains of Hydrocotyle are whitish, ry o 
beans in shape, being dk atti on one side and flat on the miht: wih a 
longitudinal furrow. The pedicels of the flowers are constricted below t ~ ca i 
apparently without any articulation at that point, nor 1s there a trace of a caly- 
us, : 
